Coordinated Intelligent Control via Epistemic Utility Theory

Epistemic utility theory is used to model autonomous intelligent agents. an epistemic system provides a mechanism for agents to characterize their knowledge corpora, options, goals, and beliefs. Levi's rule of epistemic utility provides a principle of action for decision making by comparing the informational value of rejection with the belief of correctness. Decisions are made locally and reactively, rather than globally. Coordination is implemented between agents by sharing and learning the epistemic systems of other agents. The resulting coordination model is nonhierarchichal, heterogeneous, and does not require explicit communication between agents.